Project Title

crush — The Glamourous AI Coding Agent for Your Terminal

Overview

Crush is a sophisticated AI coding agent designed to enhance productivity in your favorite terminal. It stands out with its multi-model support, flexibility to switch LLMs mid-session, and session-based context preservation. Crush is built to integrate seamlessly with your tools, code, and workflows, offering a powerful extension of your development environment.

Key Features

• Multi-Model Support: Choose from a variety of LLMs or add your own via OpenAI- or Anthropic-compatible APIs.
• Flexible LLM Switching: Maintain context while switching between LLMs during a session.
• Session-Based Context: Manage multiple work sessions and contexts per project.
• LSP-Enhanced: Utilize Language Server Protocols (LSPs) for additional context.
• Extensibility: Add capabilities via Message Content Providers (MCPs) with support for http, stdio, and sse.

Use Cases

• Developers: Use Crush to streamline coding tasks and leverage AI for code generation and enhancement.
• Data Scientists: Employ Crush for automating data analysis and model development workflows.
• DevOps Engineers: Integrate Crush into CI/CD pipelines for automated testing and deployment scripts.

Advantages

• Cross-Platform Compatibility: First-class support across macOS, Linux, Windows, and various BSD systems.
• Extensible Architecture: Easily add new capabilities and integrate with existing tools.
• User-Friendly Interface: Designed to be intuitive and easy to use, enhancing developer experience.

Limitations / Considerations

• Dependency on External APIs: Requires access to LLM APIs, which may have usage limits or costs.
• Learning Curve: There might be an initial learning curve for new users to fully leverage Crush's capabilities.

Similar / Related Projects

• Copilot by GitHub: An AI pair programmer that writes code and reduces the time spent on boilerplate tasks. Unlike Crush, it is tightly integrated with the GitHub ecosystem.
• TabNine: An AI code completion tool that offers fast and accurate suggestions. TabNine focuses on code completion, while Crush offers a broader range of AI-assisted development features.

Basic Information

• GitHub: https://github.com/charmbracelet/crush
• Stars: 12,493
• License: Unknown
• Last Commit: 2025-09-11

📊 Project Information

• Project Name: crush
• GitHub URL: https://github.com/charmbracelet/crush
• Programming Language: Go
• ⭐ Stars: 12,493
• 🍴 Forks: 616
• 📅 Created: 2025-05-21
• 🔄 Last Updated: 2025-09-11

🏷️ Project Topics

Topics: [, ]

🔗 Related Resource Links

🌐 Related Websites

• NUR

This article is automatically generated by AI based on GitHub project information and README content analysis